8 BULLETIN 1111, U. S. DEPARTMENT OF AGRICULTURE. 
superior varieties would be adopted readily, so that only small quan- 
tities of select seed would need to be distributed, which farmers or 
seedsmen would promptly increase for large-scale production. Sev- 
eral new kinds of special selections of different varieties were fur- 
nished in each distribution for several years, but the results did not 
justify the expectation that superior varieties would be readily and 
promptly utilized. It was not understood by the farmers or the 
seedsmen that the supplies of seed were limited to the stocks that were 
distributed, and no effective precautions were taken to develop or 
maintain large supplies of seed of the new varieties or the select 
strains. 
Instead of serving a useful purpose, the plan of making single 
distributions of new kinds of seed seemed only to be adding to the 
general mixture of sorts. Further experience showed that even re- 
peated distributions of seed in small quantities, of even the most 
desirable new kinds, did not result in the varieties becoming estab- 
lished in cultivation. Many farmers sent very favorable reports on 
the behavior of the new varieties, but only a few saved the seed sepa- 
rately, and very few cases were found, either among farmers or pro- 
fessional seedsmen, who on their own account maintained the isola- 
tion of the fields and separate ginning of seed long enough to reach 
the stage of commercial production. 
Thus, the method of distribution that was first projected did 
not result in establishing commercial supplies of pure seed. Several 
of the varieties that were developed and distributed in the early 
years of the cotton-breeding work were lost completely before the 
system of distribution was changed. The difficulty and uncertainty 
of securing these fundamental precautions, even among farmers 
who had active interest and were in voluntary cooperation with the 
Department of Agriculture, led finally to a clear recognition of the 
need of establishing one-variety communities. No other way has 
been suggested for maintaining and increasing stocks of pure seed on 
a scale that would provide for the general utilization of superior 
varieties. Further study and experience with one-variety com- 
munities have only confirmed and added to the reasons that were 
given in the first statement of the plan, published in 1911. 
To avoid some of the defects of the plan that had been followed 
for several years, a change in the method of conducting the distribu- 
tion of cotton seed was recommended and adopted in 1913. Smaller 
packages of seed, a quart instead of a peck, were used for the general 
distribution, but supplemented in the following year by a special 
distribution of half-bushel lots to a select number of farmers located 
in the most favorable districts for the particular varieties, as shown 
by reports furnished, supported by actual samples of bolls showing 
